Radiohalogenated proteins have been the object of extensive scientific study and promise to be useful for a variety of clinical applications, both in vitro and in vivo. For example, radioiodinated ferritin is used in an in vitro diagnostic determination of ferritin concentration in serum. Radioiodinated thyroid stimulating hormone is employed in a similar assay.
Radionuclides of halogens possess properties that make them very attractive for both diagnostic imaging and radiotherapy. For example, radioiodine as iodine-123 (T1/2=13 h, 159 keV gamma, electron capture) is nearly ideal for imaging with the current gamma cameras, and iodine-131 (T1/2=8 d, 364 keV gamma, beta particle), while producing images of lower quality, has been demonstrated to be useful in clinical radiotherapy of the thyroid. Similarly, bromine radionuclides such as bromine-75 (T1/2=1.6 h, positron) and bromine-76 (T1/2=16 h, positron) have properties that make them attractive for positron tomographic imaging, and bromine-77 (T1/2=2.4 d, several gammas, electron capture) has properties that make it attractive for radiotherapy. Other radiohalogens, such as fluorine-18 (T1/2=110 min, positron) and astatine-211 (T1/2=7.2 h, alpha particle), are also attractive candidates for radioimaging and radiotherapy.
The development of monoclonal antibodies that localize in cancerous tissue due to their high specificity and affinity for antigens on tumor cell surfaces has increased the prospect of clinical applications of radiolabeled antibodies for diagnosis and/or therapy. The high specificity of the antibodies make them desirable candidates as carrier molecules to which specific radionuclides may be attached for delivering radioactivity to a cancer site.
Other proteins, protein fragments, modified proteins, and peptides that tend to concentrate in an organ or diseased tissue are likewise candidates to be radiohalogenated. For example, radiohalogenated fibrinogen could be used to localize deep vein thrombosis by in vivo imaging. Disease-altered uptake of pituitary and other peptide hormones could be monitored in a similar manner.
Unfortunately, there are presently no routine clinical diagnostic or therapeutic applications of radiohalogen labeled antibodies for use in vivo. Direct radiohalogen labeling of antibodies and other proteins has proved to be difficult. Antibodies exhibit varying sensitivities to radiolabeling reaction conditions, and the oxidizing reaction conditions necessary for standard radiohalogenations are particularly deleterious. Direct radioiodination of proteins has become routine, but very often a measurable reduction of biological activity of the protein results. The stability of the attached radiolabel can also vary. For example, the loss of radioiodine from antibodies has been found to be as high as 50% in 24 hours for some labeled antibodies. Radiobrominations require even stronger oxidizing reaction conditions than radioiodinations, and attempts to radiobrominate proteins directly have met with little success unless expensive and difficult to obtain enzymes are used as oxidants. Furthermore, direct radiohalogenation of proteins occurs primarily at tyrosyl residues, and the activated phenol ring of tyrosine contributes to an inherent electronic instability of the resultant ortho-substituted radiohalogen label. The radiohalogen label is also subject to steric hindrance effects and may in addition be available to deiodinase enzymes which catabolize the structurally similar thyroid hormones, e.g., thyroxine.
One approach that circumvents subjecting proteins to the harsh reaction conditions necessary for direct radiohalogenations is the use of small molecules that can be radiolabeled in a separate reaction vessel and subsequently coupled to proteins under mild reaction conditions. This approach is the basis of the commercially available Bolton-Hunter reagent, N-succinimidyl-3-(4-hydroxyphenyl)propionate. Moderate radiolabeling yields are thereby obtained with radioiodine (35-60% yields of labeled proteins), but the stability of the radioiodine label suffers from the same problems as described for the chemically similar radioiodinated tyrosyl residues. Similarly, the commercially available Wood's reagent, methyl-4-hydroxybenzimidate, can be radioiodinated prior to attachment to proteins. However, the radioiodinated product is also plagued with the inherent instability of the ortho-iodinated phenol. Even though these reagents do not yield as stable as radiolabel as desirable, they have been extensively used for radioiodination because little deactivation of the protein results from their use.
The phenolic ring is employed in both the Bolton-Hunter and Wood's reagents because an activated aromatic ring is required in order to introduce high specific activity radioiodine into these molecules. It would be very desirable to be able to introduce radiohalogens into other small molecules so that the radiolabel would be more stably attached.
Recent reports in the literature describe the use of organometallic intermediates to introduce high specific activity radiohalogens into vinyl positions on alkene moieties. For example, vinylstannanes have been reportedly used to radiohalogenate steroid in the 17 position (Hanson, R. N., et al., J. Nucl. Med. 23: 431-436, 1982), and to label sugars (Goodman, M. M., et al., in Sixth International Symposium on Radiopharmaceutical Chemistry, Boston, MA, June 29-July 3, 1986, paper number 106). Use of vinyl boronic acids in the radiohalogen labeling of steroids has also been reported. Kabalka, G. W., et al., Synth. Commun. 11: 247-251, 1981; Kabalka, G. W., et al., Applications of Nuclear and Radiochemistry, Lambrecht, R. M., et al., Eds., Pergamon Press, Newark, N.J., Chapter 17, pp. 197-203, 1981.